Diet selection, intake and weight gain of dorper sheep fed on selected grasses in a cafeteria system in Machakos County, Kenya

Livestock is a key asset and a primary livelihood resource for rural households in most parts of the world and accounts for nearly 95 per cent of family income in the Arid and Semi-Arid Lands (ASALs) in Kenya. Despite high economic importance of livestock, inadequate nutrition results to low livestock productivity in Kenya. Ruminant feeding in the ASALs is mainly based on the exploitation of range feed resources. Grazing ruminants forage on different plant species with varying levels of nutrient and toxins. Despite these complex grazing situations, sheep can select a suitable diet and regulate intake of toxic substances. There is therefore a need understand diet selection and its role in the nutrition of sheep. A study to assess diet selection, intake and live weight gain of sheep fed on different grasses was conducted at Machakos Agriculture Training Centre (ATC) in two experiments. Five sheep housed in individual pens of size (3 x 3m) were fed in a cafeteria system with six grasses namely Brachiaria decumbens cv. Basilisk, Chloris gayana, Cenchrus ciliaris, Chloris roxburghiana, Enteropogon macrostachyus and Eragrostis superba were used during experiment 1. Data on dry matter intake (DMI), feeding time and number of visits were recorded for 5 consecutive days. Selectivity index (SI) for each grass was calculated from intake data. The grasses were analyzed for chemical composition according to Association of Official Analytical Chemist (AOAC) (1990). Analysis of variance (ANOVA) was conducted using the SAS, 2000 model to determine the effect of grass species on diet selection. Four grasses and sixteen Dorper sheep aged between 12 and 18 months and weighing between 18 and 29 Kgs were used in experiment 2. The sheep were grouped into four groups balanced for age and weight. Each group was randomly allocated one dietary treatment. Mineral licks and water were provided ad libitum throughout the study. Data on intake and weight gain were computed for 42 days and analyzed by General Linear Model (GLM) procedure of SAS 2000. Results of experiment 1 showed that number of visits, feeding time, and the selectivity index (SI) were higher (p